This paper presents instruct-imagen, a model that tackles heterogeneous image generation tasks and generalizes across unseen tasks. We introduce *multi-modal instruction* for image generation, a task representation articulating a range of generation intents with precision. It uses natural language to amalgamate disparate modalities (e.g., text, edge, style, subject, etc.), such that abundant generation intents can be standardized in a uniform format. We then build instruct-imagen by fine-tuning a pre-trained text-to-image diffusion model with a two-stage framework. First, we adapt the model using the retrieval-augmented training, to enhance model's capabilities to ground its generation on external multimodal context. Subsequently, we fine-tune the adapted model on diverse image generation tasks that requires vision-language understanding (e.g., subject-driven generation, etc.), each paired with a multi-modal instruction encapsulating the task's essence. Human evaluation on various image generation datasets reveals that instruct-imagen matches or surpasses prior task-specific models in-domain and demonstrates promising generalization to unseen and more complex tasks.

Disease dynamics, human mobility, and public policies co-evolve during a pandemic such as COVID-19. Understanding dynamic human mobility changes and spatial interaction patterns are crucial for understanding and forecasting COVID-19 dynamics. We introduce a novel graph-based neural network(GNN) to incorporate global aggregated mobility flows for a better understanding of the impact of human mobility on COVID-19 dynamics as well as better forecasting of disease dynamics. We propose a recurrent message passing graph neural network that embeds spatio-temporal disease dynamics and human mobility dynamics for daily state-level new confirmed cases forecasting. This work represents one of the early papers on the use of GNNs to forecast COVID-19 incidence dynamics and our methods are competitive to existing methods. We show that the spatial and temporal dynamic mobility graph leveraged by the graph neural network enables better long-term forecasting performance compared to baselines.

In this paper, we present a new approach to time series forecasting. Time series data are prevalent in many scientific and engineering disciplines. Time series forecasting is a crucial task in modeling time series data, and is an important area of machine learning. In this work we developed a novel method that employs Transformer-based machine learning models to forecast time series data. This approach works by leveraging self-attention mechanisms to learn complex patterns and dynamics from time series data. Moreover, it is a generic framework and can be applied to univariate and multivariate time series data, as well as time series embeddings. Using influenza-like illness (ILI) forecasting as a case study, we show that the forecasting results produced by our approach are favorably comparable to the state-of-the-art.

As rates of new coronavirus disease 2019 (COVID-19) cases decline across Europe owing to nonpharmaceutical interventions such as social distancing policies and lockdown measures, countries require guidance on how to ease restrictions while minimizing the risk of resurgent outbreaks. We use mobility and case data to quantify how coordinated exit strategies could delay continental resurgence and limit community transmission of COVID-19. We find that a resurgent continental epidemic could occur as many as 5 weeks earlier when well-connected countries with stringent existing interventions end their interventions prematurely. Further, we find that appropriate coordination can greatly improve the likelihood of eliminating community transmission throughout Europe. In particular, synchronizing intermittent lockdowns across Europe means that half as many lockdown periods would be required to end continent-wide community transmission.

In this work, we examine a novel forecasting approach for COVID-19 case prediction that uses Graph Neural Networks and mobility data. In contrast to existing time series forecasting models, the proposed approach learns from a single large-scale spatio-temporal graph, where nodes represent the region-level human mobility, spatial edges represent the human mobility based inter-region connectivity, and temporal edges represent node features through time. We evaluate this approach on the US county level COVID-19 dataset, and demonstrate that the rich spatial and temporal information leveraged by the graph neural network allows the model to learn complex dynamics. We show a 6% reduction of RMSLE and an absolute Pearson Correlation improvement from 0.9978 to 0.998 compared to the best performing baseline models. This novel source of information combined with graph based deep learning approaches can be a powerful tool to understand the spread and evolution of COVID-19. We encourage others to further develop a novel modeling paradigm for infectious disease based on GNNs and high resolution mobility data.

AbstractIn order to prepare for and control the continued spread of the COVID-19 pandemic while minimizing its economic impact, the world needs to be able to estimate and predict COVID-19’s spread. Unfortunately, we cannot directly observe the prevalence or growth rate of COVID-19; these must be inferred using some kind of model. We propose a hierarchical Bayesian extension to the classic susceptible-exposed-infected-removed (SEIR) compartmental model that adds compartments to account for isolation and death and allows the infection rate to vary as a function of both mobility data collected from mobile phones and a latent time-varying factor that accounts for changes in behavior not captured by mobility data. Since confirmed-case data is unreliable, we infer the model’s parameters conditioned on deaths data. We replace the exponential-waiting-time assumption of classic compartmental models with Erlang distributions, which allows for a more realistic model of the long lag between exposure and death. The mobility data gives us a leading indicator that can quickly detect changes in the pandemic’s local growth rate and forecast changes in death rates weeks ahead of time. This is an analysis of observational data, so any causal interpretations of the model’s inferences should be treated as suggestive at best; nonetheless, the model’s inferred relationship between different kinds of trips and the infection rate do suggest some possible hypotheses about what kinds of activities might contribute most to COVID-19’s spread.

AbstractTimely interventions and early preparedness of healthcare resources are crucial measures to tackle the COVID-19 disease. To aid these efforts, we developed the Mobility-Augmented SEIR model (MA-SEIR) that leverages Google’s aggregate and anonymized mobility data to augment classic compartmental models. We show in a retrospective analysis how this method can be applied at an early stage in the COVID-19 epidemic to forecast its subsequent spread and onset in different geographic regions, with minimal parameterization of the model. This provides insight into the role of near real-time aggregate mobility data in disease spread modeling by quantifying substantial changes in how populations move both locally and globally. These changes would be otherwise very hard to capture using less timely data.

This work quantifies mobility changes observed during the different phases of the pandemic world-wide at multiple resolutions -- county, state, country -- using an anonymized aggregate mobility map that captures population flows between geographic cells of size 5 km 2 . As we overlay the global mobility map with epidemic incidence curves and dates of government interventions, we observe that as case counts rose, mobility fell and has since then seen a slow but steady increase in flows. Further, in order to understand mixing within a region, we propose a new metric to quantify the effect of social distancing on the basis of mobility.Taking two very different countries sampled from the global spectrum, We analyze in detail the mobility patterns of the United States (US) and India. We then carry out a counterfactual analysis of delaying the lockdown and show that a one week delay would have doubled the reported number of cases in the US and India. Finally, we quantify the effect of college students returning back to school for the fall semester on COVID-19 dynamics in the surrounding community. We employ the data from a recent university outbreak (reported on August 16, 2020) to infer possible R eff values and mobility flows combined with daily prevalence data and census data to obtain an estimate of new cases that might arrive on a college campus. We find that maintaining social distancing at existing levels would be effective in mitigating the extra seeding of cases. However, potential behavioral change and increased social interaction amongst students (30% increase in R eff ) along with extra seeding can increase the number of cases by 20% over a period of one month in the encompassing county. To our knowledge, this work is the first to model in near real-time, the interplay of human mobility, epidemic dynamics and public policies across multiple spatial resolutions and at a global scale.

We report the results of a computational, atomistic electrodynamics study of the effects of electromagnetic waves on the mechanical properties and specifically the Young’s modulus of silver nanowires. We find that the Young’s modulus of the nanowires is strongly dependent on the optical excitation energy, with a peak enhancement occurring at the localized surface plasmon resonance frequency. When the nanowire is excited at the plasmon resonance frequency, the Young’s modulus is found to increase linearly with increasing nanowire aspect ratio, with a stiffening of nearly 15% for a 2 nm cross section silver nanowire with an aspect ratio of 3.5. Furthermore, our results suggest that this plasmon resonance-induced stiffening is stronger for larger diameter nanowires for a given aspect ratio. Our study demonstrates a novel approach to actively tailoring and enhancing the mechanical properties of metal nanowires.

We present a computational, atomistic study of electric field effects on the Young's modulus of metal nanowires. The simulations are electromechanically coupled, where the mechanical forces on the atoms are obtained from realistic embedded atom method potentials, and where the electrostatic forces on the atoms are obtained using a point dipole electrostatic model that is modified to account for the different polarizability and bonding environment of surface atoms. By considering three different nanowire axial orientations ([100], [110] and [111]) of varying cross sectional sizes and aspect ratios, we find that the Young's modulus of the nanowires differs from that predicted for the purely mechanical case due to the elimination of nonlinear elastic stiffening or softening effects due to the electric field-induced positive relaxation strain relative to the relaxed mechanical configuration. We further find that [100] nanowires are most sensitive to the applied electric field, with Young's moduli that can be increased more than 20% with increasing aspect ratio. Finally, while the orientation of the transverse surfaces does impact the Young's modulus of the nanowires under applied electric field, the key factor controlling the magnitude of the stiffness change of the nanowires is the distance between atomic planes along the axial direction of the nanowire bulk.

The time-to-digital converter(TDC) aims to mark an accurate timestamp at the time of input signal comes. The Multi-phase Clock sampling method is an usual way to map the TDC into an FPGA. Traditionally, this method provides a medium accuracy and low resources occupation. In this paper, we present a new architecture of TDC base on the 2-ISERDES in the SelectIO, rather than utilizing the Slice resources by the old way. The ISERDESes based TDC is equivalent to a 8 equidistant phase-shifted clocks TDC, with maximum clock frequency 900MHz. The least significant bit(LSB) is 139ps, which is 445% better than traditional architecture.

We present in this work the notion of using elastic strain engineering to reduce the intrinsic losses in a metal for subwavelength optical signal processing. By using a simple, analytical waveguide model, we demonstrate that application of uniaxial tensile strains below the yield strain of gold nanowires results in substantial increases of more than 70% in the surface plasmon polariton propagation lengths at optical frequencies. The enhancement is primarily due to a reduction in the core electron density, and is found to be size-independent for a wide range of nanowire diameters, while exhibiting a linear dependence on the applied tensile strain.

We present a computational, atomistic electrodynamics investigation of the effects of planar defects on the optical properties of silver nanocubes, where the planar defects we considered are different surface orientations, twins, partial dislocations, and full dislocations. We find that for nanocubes smaller than about 3 nm, the optical response is very sensitive to the specific surface structure resulting from the defects. However, the sensitivity, as measured by shifts in the plasmon resonance wavelength, is strongly reduced at larger sizes because of the decreasing importance of surface effects even when the majority of the atomic deformation due to the crystal defects is contained within the interior of the nanocube. Overall, this study suggests that the effects of individual crystalline defects on the optical properties of nanostructures can be safely ignored for nanostructure sizes larger than about 5 nm.

We study the validity of the recently proposed universal plasmon ruler in the present work using a combination of numerical techniques based on the finite difference time domain (FDTD) method, and semianalytical theories based on the coupled dipole approximation. By incorporating nonlocal effects for closely spaced two-dimensional gold nanocylinder dimers, we find using both the FDTD and semianalytical approaches that the universal plasmon ruler of Jain et al.(1) is not applicable for gold nanocylinder dimers with diameters smaller than about 20 nm. The nonlocal effects are also found to strongly reduce the electric field enhancements at very small gap distances. Taken together with previous results,(2) we are able to establish the valid size range for the universal plasmon ruler of gold: 20 nm diameter metal nanostructure dimers at the smaller end, and 70 nm diameter metal nanostructure dimers at the larger end.

We have utilized the discrete dipole approximation to study the localized surface plasmon resonance in infinite, periodic two-dimensional arrays of gold nanospheres with the nanospheres arranged according to the {100} face of an fcc crystal. Specifically, we have performed a systematic study of the sensitivity of both the plasmon resonance wavelength shift and extinction properties considering nanosphere diameters ranging from 20 to 100 nm, and for nanosphere gap distances ranging from 0.5 to 6 times the nanosphere diameter. In doing so, we find that the same universal decay length of the plasmon resonance wavelength shift of about 0.2 units of the nanosphere size that was previously found by Jain et al.(1) for nanoparticle dimers is also operant for two-dimensional arrays. However, we also find that the universality of the plasmon ruler is only valid for arrays with nanospheres smaller than a critical nanosphere diameter of about 70 nm, whereas for larger nanosphere diameters, a decrease in the extinction efficiency as the gap distance decreases and a reduction in the decay constant are observed. Both of these size-dependent optical responses are qualitatively interpreted using a semianalytical coupled dipole approximation that accounts for structural retardation due to the geometric arrangement of the nanospheres, as well as single sphere retardation due to both dynamic depolarization and radiative damping effects. Using the semianalytical theory, we find that the size dependence is primarily due to the effects of dynamic depolarization and structural retardation, which reduces the coupling strength, changes the extinction efficiency trend, and also reduces the decay constant of the plasmon ruler equation for larger diameter nanospheres; similar results were found for infinite, 2D arrays of nanospheres in hexagonal and simple cubic arrangements. Finally, the semianalytical theory is utilized to predict a size dependence of the plasmon ruler for dimers starting at the same critical diameter of 70 nm. However, we find that the size effect is weaker for dimers than for the array case due to the significant reduction in structural retardation for dimers as compared to the array case.